Pain can be modulated by explicit beliefs about treatments, prior experience and learning, interpersonal processes that support the patient-provider relationship, and contextual factors related to the treatment environment. In this project, we systematically investigate the neural and psychological mechanisms that mediate the effects of these factors on acute pain. We focus on expectations, attention, emotion, conditioning/associative learning, and social factors. Our experiments principally use functional magnetic resonance imaging (fMRI) and psychophysiological measurements, as well as behavioral assays and self-reports. We are examining the effects of different types of pain-related expectations on decisions about pain as well as responses in the brain and body, and we are comparing acute pain with other hedonic and perceptual processes (e.g. pleasant and unpleasant taste perception). This will allow us to distinguish processes that are unique to pain perception from those that are not specific to pain, such as processes involved in perception and decision-making across domains. This was the fourth year of the Section on Affective Neuroscience and Pain, and the lab continued to grow, adding one postdoctoral fellow, three new postbaccalaureate research assistants, and a data analyst. We also had three fellows complete their time in our lab successfully and move on to new positions: One postdoctoral fellow (DM) received a faculty position and two former post baccalaureate research assistants were accepted into PhD programs (Clinical Psychology and Behavioral and Cognitive Neuroscience). We continued to make strong progress on our thematic human subjects protocol Neural and psychological mechanisms of pain perception. The protocol includes five sub-studies designed to a) isolate different aspects of pain modulation, b) compare acute pain modalities (e.g., thermal pain versus shock-induced pain), and c) compare and contrast pain with other hedonic and perceptual domains (e.g., taste). In all studies, we measure decisions about pain experience (self-report) as well as neural and physiological responses to noxious stimuli that cause pain. During analysis, we combine computational modeling with advanced neuroimaging analyses to isolate the neural and psychological mechanisms that mediate the effects of expectations, attention, and emotion on subjective pain. Our protocol requires all participants to go through an initial calibration session, following screening. Participants complete questionnaires, and then undergo a procedure that measures pain ratings in response to noxious heat stimuli and determines each participants pain threshold and tolerance. 233 individuals have completed this procedure to date (50 participants so far during Year 4 of the protocol, FY17-18). We published one manuscript from this protocol in the past fiscal year (Mischkowski et al., Pain, 2018), which was selected as an Editors pick for the journal. We analyzed the relationship between temperature, pain, and autonomic responses (skin conductance and pupil dilation), and found that autonomic responses to objective stimuli depend on whether a response is classified as painful or not, and likewise the intensity of the pain drives the bodys response, rather than the objective stimulus temperature. This paper has been well-received and was selected as an Editors pick at the journal. We are currently preparing two manuscripts for publication from a set of studies that were completed in FY17, which examine the effects of classical conditioning and instructed knowledge on pain and reversal learning. Forty healthy volunteers successfully completed the fMRI task and we are currently analyzing fMRI data from these participants. We are currently analyzing fMRI data, with the help of our new fMRI data analyst and in collaboration with scientists in NIMH. In addition, we are preparing a manuscript for submission (Atlas et al., Invited submission, Journal of Abnormal Psychiatry) that measures the influence of anxiety on pain and pain-related learning by combining the fMRI subjects data with data from 23 participants who completed the task outside the scanner. We find dissociations between cue-based expectancy effects on pain reports and skin conductance responses, and that both types of responses are related to individual differences in state anxiety within our healthy volunteers. We are currently following up on these findings through a collaboration with Dr. Daniel Pine (NIMH) to test reversal learning and pain in clinically anxious individuals. We also completed a second fMRI sub-study that tests whether cue-based expectancies and treatment-based expectancies (i.e. placebo responses) modulate pain through dissociable pathways. This study crosses conditioned cues with a placebo analgesia manipulation. FMRI scanning began in the spring of 2017, and was completed in June of 2018, with 40 participants who successfully completed the task. Preliminary analyses indicate strong cue-based expectancy effects, but large variability in placebo effects across the sample. We are now measuring the brain pathways that support each of these effects, and expect to submit a manuscript in FY19. This project is led by a postdoc in the lab, Elizabeth Necka. We are currently in the process of collecting fMRI data for a third sub-study that measures the relationship between pain and taste and how expectations modulate perceptions across domains. We worked with the Section on Instrumentation to build an fMRI-compatible gustometer (taste-delivery device) that can deliver juices in the fMRI scanner and began a completed a pilot sub-study in the NCCIH mock scanner that administers heat along with pleasant and unpleasant liquid tastants in the beginning of FY18. In our pilot sample of 18 subjects (6 who received salt solutions, 6 who received sugar solutions, 6 who received thermal pain). We found robust cue-based pain modulation in each group, and conducted power analyses that indicated that 20 subjects per group would provide adequate power in the fMRI study. We are now well underway with the fMRI version of the study: 11 participants have completed all visits, and 8 participants have completed the first visit to select temperatures and tastants prior to the fMRI study. The overall goal of this project is to isolate domain-specific as well as domain-general mechanisms that underlie expectancy, affective learning, and perception. This project is being led by our new postdoc, Inseon Lee, along with two new postbaccalaureate students. Finally, we also completed behavioral testing on a sub-study that measures the relationship between attention, expectations, and learning. 18 participants completed the behavioral sub-study, but our hypothesized interactions between expectations and learning were not observed, therefore we decided not to adapt this paradigm for the imaging environment. We may return to this question in the future by developing a new paradigm to manipulate attention.